Proceedings of the 5th International Conference on Big Data and Internet of Things (Lecture Notes in Networks and Systems, 489)

This book is a collection of papers in the research area of big data, cloud computing, cybersecurity, machine learning, deep learning, e-learning, Internet of Things, reinforcement learning, information system, social media and natural language processing. This book includes papers presented at the 5th International Conference on Big Data Cloud and Internet of Things, BDIoT 2021 during March 17–18, 2021, at ENSIAS, Mohammed V University in Rabat, Morocco. Contents Big Data and Cloud Computing Toward an Automatic Assistance Framework for the Selection and Configuration of Machine Learning Based Data Analytics Solutions in Industry 4.0 1 Introduction 2 Related Works 3 Automated Machine Learning 3.1 Meta-learning for Automatic Algorithms Selection and Configuration 4 Automated Machine Learning in Industry 4.0 5 Framework and Methodology 5.1 Learning Phase 5.2 Recommendation Phase 5.3 Prototypical Implementation 6 Conclusion References New Deep Learning Architecture for Improving the Accuracy and the Inference Time of Traffic Signs Classification in Intelligent Vehicles 1 Introduction 2 Big Data Challenges 2.1 Machine Learning Approaches 2.2 Deep Learning Approaches 3 Traffic Signs Recognition 3.1 Traffic Signs Detection 3.2 Traffic Signs Classification 4 Proposed Architecture 5 Obtained Results 5.1 Training Results 5.2 Testing Results 6 Conclusion 7 Further Work References Implicit JSON Schema Versioning Triggered by Temporal Updates to JSON-Based Big Data in the τJSchema Framework 1 Introduction 2 Background 2.1 The τJSchema Framework 2.2 Temporal JSON Data Updates 3 The Proposed Approach 4 Illustrative Example 5 Conclusion and Future Work References A Fuzzy Meta Model for Adjusting Ant Colony System Parameters 1 Introduction 2 Background 2.1 Ant Colony System 2.2 Fuzzy Logic Controller 3 Proposed Method 4 Experimental Results 5 Statistical Test 6 Conclusion and Future Work References Ontology Engineering Methodologies: State of the Art 1 Introduction 2 Ontology Development Methods Review 2.1 Methodologies for Building Ontologies 2.2 Summary Analysis of Methodologies 3 Conclusion References Conception of an Automatic Decision Support Platform Based on Cross-Sorting Methods and the Fuzzy Logic for General Use 1 Introduction 2 The Calculation Algorithm 3 The Consistency Index 4 Optimization Algorithm 4.1 Literature Review 4.2 Simulated Annealing: Algorithm Description 5 Simulation and Results 6 Conclusion References Cyber Security Classification of URLs Using N-gram Machine Learning Approach 1 Introduction 2 Related Work 3 Methodology 3.1 Step1: Data Collection and Processing 3.2 Step2: N-gram Approach and Scaling 3.3 Performance Evaluation 4 Implementation and Results 5 Conclusion References Denial of Service Attack Detection in Wireless Sensor Networks and Software Defined Wireless Sensor Networks: A Brief Review 1 Introduction 2 Background 2.1 Wireless Sensor Networks (WSN) 2.2 Software Defined Wireless Sensor Network (SDWSN) 2.3 Denial of Service Attack 3 Existing Solutions for DoS Attack Detection in WSN and SDWSN 3.1 Solutions for WSN 3.2 Solutions for SDWSN 4 Discussion 5 Conclusion References An RGB Image Encryption Algorithm Based on Clifford Attractors with a Bilinear Transformation 1 Introduction 2 Clifford Attractor 3 Proposed Approach 4 Statistical Analyzes 4.1 Histogram 4.2 Correlation Coefficient of Two Adjacent Pixels 4.3 Differential Attacks 4.4 Information Entropy Analysis 5 Conclusion References An Overview of Security in Vehicular Ad Hoc Networks 1 Introduction 2 Vehicular Ad Hoc Networks ``VANETs'' 3 Security of Vehicular Ad Hoc Networks 4 Discussion 5 Conclusion References Adaptive Approach of Credit Card Fraud Detection Using Machine Learning Algorithms 1 Introduction 2 Related work 2.1 Logistic Regression 2.2 Naïve Bayes 2.3 Random Forest 2.4 Multilayer Perceptron 3 Experimental result 3.1 Dataset Description 3.2 Google Colaboratory 3.3 Performance Metrics 3.4 Obtained Result 4 Adaptive approach 5 Conclusion References Homomorphic Method Additive Using Pailler and Multiplicative Based on RSA in Integers Numbers 1 Introduction 2 Related Works 3 Homomorphic Encryption System 3.1 Partially Homomorphic Cryptographic Scheme 4 Proposed Method 4.1 Homomorphic Additive and Multiplicative Encryption/Decryption Algorithm 5 Experimentation 5.1 Paillier Encryption/Decryption 5.2 Multiplication Based on RSA 5.3 Application 6 Conclusion References Deep Learning Hybrid Deep Learning Models for Diabetic Retinopathy Classification 1 Introduction 2 Motivation: Diabetic Retinopathy 2.1 Diabetic Retinopathy Characteristics 2.2 Diabetic Retinopathy Classification 3 Related Works 3.1 Traditional Vision Methods 3.2 Artificial Intelligence Methods 4 Diabetic Retinopathy Recognition in Big Data Environment 4.1 Database Description 4.2 Hybrid CNN-KNN/SVM Model 4.3 Results and Discussion 5 Conclusion and Future Work References Detection of Appliance-Level Abnormal Energy Consumption in Buildings Using Autoencoders and Micro-moments 1 Introduction 2 Related Works 3 Proposed System 3.1 Preprocessing 3.2 Autoencoders 3.3 Micro-moment Clustering 4 Experimental Results 5 Conclusion References Predicting the Mode of Transport from GPS Trajectories 1 Introduction 2 Methodology and Processing the Input 2.1 Preparing the Data 3 Experiments Details and Results 3.1 Data Description 3.2 Experiments and Performances 3.3 Comparison and Discussion 4 Conclusion References Application of Artificial Intelligence to X-ray Image-Based Coronavirus Diseases (COVID-19) for Automatic Detection 1 Introduction 2 Related Work 3 Proposed Image-based COVID-19 Disease Detection Model 3.1 Dataset 3.2 Model Architecture Overview 4 Experimental Results and Discussion 4.1 Experimental Details 4.2 Results 4.3 Discussion 5 Conclusion References E-Learning Personalization Between Pedagogy and Adaptive Hypermedia System 1 Introduction 2 Scientific and Problematic Context 2.1 An overview of the Concept of Personalization 2.2 Personalized Educational Hypermedia Systems 2.3 The Pedagogical Practices of Personalization 2.4 Learning Styles 3 Conclusion References Collaboration in Adaptive E Learning 1 Introduction 2 Collaborative Learning 3 Adaptive Learning 4 Structural Pedagogy 4.1 Educational Strategies 4.2 Educational Style 5 Collaborative Work in Adaptive Learning 6 Conclusions References A Review of the State of Higher Education in MOROCCO at the Time of Covid-19 1 Introduction 2 Review of Distance Education 2.1 History 2.2 Types of Distance Learning 2.3 Evolution of Distance Learning Platforms 3 Current Situation of the Use of ICT in Moroccan Higher Education 3.1 Analysis of the Use of ICT in Moroccan Higher Education 4 Comparative Study of the Various Current EAD Platforms 4.1 Comparative Technical and Educational Study of the Selected Platforms 4.2 Comparative Analysis of Platforms 4.3 Challenges in the Development of an E-learning System 5 Conclusion References Internet of Things Digital Twin-Driven Approach for Smart Industrial Product Design 1 Introduction 2 Digital Twin: Concept and Applications 2.1 Concept of DT 2.2 Digital Twin Applications 3 Big Data Analytics, IoT and Digital Twin Challenges in the Area of Industrial Product Development 4 The Proposed Digital Twin-Driven Approach for Smart Product Design 4.1 Motivations 4.2 The Approach Outlines 5 Case of Study 6 Conclusion References Towards a Generic Architecture of Context-Aware and Intentional System 1 Introduction 2 Motivating Scenario 3 Background and Related Work 3.1 Context-Awareness 3.2 Intentionality 4 Contribution 4.1 Intention and Context Relation 4.2 Proposal for a Preliminary Architecture 4.3 The Proposed Intentional Model 5 Conclusion References A Smart Healthcare Imbalanced Classes Model Using Multi Conditional-Task GAN 1 Introduction 2 Related Work 2.1 Class Imbalance Problem 2.2 Generative Adversarial Networks 2.3 Conditional Generative Adversarial Networks 2.4 Multi-task Generative Adversarial Networks 3 The Proposed Solution 3.1 Conditional GAN (cGAN) 3.2 Multi-Task GAN (MTGAN) 3.3 Learning General Data Distribution 3.4 Anomaly Detection 4 Case Study 4.1 Smart Healthcare Domain 4.2 System Description 4.3 System Problem 4.4 Smart Healthcare Application 5 Conclusion References Machine Learning Prediction of Risks in Intelligent Transport Systems 1 Introduction 2 Related Work 3 Comparative Study and Discussion 3.1 Comparative Study 3.2 Discussion 4 Experiments 4.1 Dataset and Preprocessing 4.2 Classification Models and Performance Measurement 5 Conclusions References A Systematic Literature Review of Machine Learning Applications in Software Engineering 1 Introduction 2 Research Methodology 2.1 Research Questions 2.2 Inclusion and Exclusion Criteria 2.3 Search Strategy 2.4 Study Selection and Data Extraction 2.5 Selected Studies Overview 3 Results 3.1 RQ1: What SE Research Topics have been Addressed in the Selected Studies? 3.2 RQ2: What ML Algorithms have been used in the Selected Studies? 3.3 RQ3: What are the Main Outcomes of the Selected Studies? 4 Conclusion References Machine Learning for Used Cars Price Prediction: Moroccan Use Case 1 Introduction 2 Literature Review and Comparison 3 Materials and Methods 3.1 Background Theory 3.2 Methodology 4 Experiments and Results 4.1 Hyper Parameter Tuning 4.2 Evaluation Metric 4.3 Results 4.4 Model Deployment 5 Conclusion References Improving Team Performance by Using Clustering and Features Reduction Techniques 1 Introduction 2 Big Data, Machine Learning and Human Resources Management 2.1 Big Data 2.2 Machine Learning 2.3 Human Resources Management 3 Related Works 4 Improving Team Performance By Using Clustering and Features Reduction Techniques 4.1 Presentation of the Data 4.2 Model Building 5 Conclusion References Home Automation and Machine Learning Models for Health Monitoring 1 Introduction 2 Home Automation and Health Monitoring 2.1 Home Automation Technologies for Health Monitoring 2.2 The Proposed Home Automation Architecture for Health Monitoring 3 Machine Learning for Health Monitoring 3.1 Machine Learning Applications 3.2 Machine Learning for Health Monitoring 4 Case Study: Heart Disease Monitoring Using Machine Learning 4.1 Machine Learning Application Results 4.2 Results Discussion 5 Conclusion References A Study of Machine Learning Based Approach for Hotels' Matching 1 Introduction 2 Related Work 3 Dataset 4 Proposed Solution 4.1 Data Pre-processing 4.2 Similarity Measuring 4.3 Machine Learning Model 4.4 Evaluation of the Model 5 Results and Discussion 6 Conclusion References Natural Processing Language Empirical Study: What is the Best N-Gram Graphical Indexing Technique 1 Introduction 2 Indexing Techniques Based on n-gram Graphs 2.1 History 2.2 N-gram Graph Concepts 2.3 N-gram Graph Process 2.4 K-core Graph Process 2.5 N-gram Graph Indexing Algorithm 3 Empirical Study 3.1 Data Set 3.2 Character N-gram Graph 3.3 Word N-gram Graph 4 Discussion 5 Conclusion and Perspectives References Transformer Model and Convolutional Neural Networks (CNNs) for Arabic to English Machine Translation 1 Introduction 2 Related Works 3 Our Proposed Approach 4 Experiments 4.1 Dataset and Preprocessing 4.2 Hyperparameters 4.3 Experimental Details and Results 5 Conclusion References An Overview of Word Embedding Models Evaluation for Arabic Sentiment Analysis 1 Introduction 2 Arabic Sentiment Analysis 2.1 Arabic Sentiment Analysis Levels 2.2 Arabic Sentiment Analysis Feature Types 3 Word Embeddings in Arabic Sentiment Analysis 3.1 Word2Vec 3.2 Global Vectors (GloVe) 3.3 FastText 3.4 Elmo 3.5 Bert 4 Results and Discussion 5 Conclusion and Future Work References A Hybrid Learning Approach for Text Classification Using Natural Language Processing 1 Introduction 2 Related Work 3 Text Classification Process 3.1 Preprocessing 3.2 Text Representation 4 Classification Techniques 4.1 Naive Bayesian Algorithm 4.2 Logistic Regression Algorithm 4.3 The Proposed Hybrid Approach 5 Performance Evaluation 5.1 Classification Dataset 5.2 Results and Discussion 6 Conclusion References Automatic Key-Phrase Extraction: Empirical Study of Graph-Based Methods 1 Introduction 2 AKE Process 2.1 Preprocessing 2.2 Candidate Key-Phrases 2.3 Key-Phrases Extraction 2.4 The Evaluation 3 Graph-Based Key-Phrases Extraction 3.1 TextRank Approach 3.2 ExpandRank Approach 3.3 TopicRank approach 3.4 K-Core Approach 3.5 WordAttractionRank Approach 3.6 PositionRank Approach 3.7 Multipartite Graph Approach 3.8 RaKUn Approach 4 Evaluation 4.1 Datasets 4.2 Evaluation Metric 4.3 Results 4.4 Discussion 5 Conclusion References Sentiment Analysis of Moroccan Dialect Using Deep Learning 1 Introduction 2 Sentiment Analysis 2.1 Moroccan Dialect 2.2 Sentiment Analysis Techniques 3 Related Work 4 Proposed Approach 4.1 Collecting Data 4.2 Data Preprocessing and Cleaning 4.3 Word Representation 5 Build Vocabulary for Our Comments: 5.1 Convolutional Neural Network 6 Experimental Results 7 Conclusion References Reinforcement Learning Policy Gradient for Arabic to English Neural Machine Translation 1 Introduction 2 Background 2.1 Neural Machine Translation 2.2 Performance Evaluation of Machine Translation 2.3 Policy Gradient 3 Policy Gradient Training for NMT 4 Experiments 4.1 Dataset 4.2 Experimental Settings 4.3 Results 5 Conclusion A Proof of Equation11 References Dilemma Game at Unsignalized Intersection with Reinforcement Learning 1 Introduction 2 Traffic Model and Dilemma Game 2.1 Model Structure and Moving Rules 2.2 Dilemma Game at the Intersection 3 Q-Learning Algorithm 3.1 Structure of Q-Learning 3.2 State-Action Pairs 3.3 Rewards and Penalties Functions 3.4 Simulation Method and Performance Metrics 4 Results and Discussion 5 Conclusions References Toward a Self-adaptive Supply Chains: L-SCOR Implementation Proposal, and Case Studies Methodology Proposal 1 Introduction 1.1 Self-adaptive Systems 1.2 L-SCOR: An Extension of SCOR for a Self-adaptive Supply Chain 2 L-SCOR Implementation 2.1 Q-Learning 2.2 Web Services 3 Case Studies Methodology 4 Conclusion References The Retirement Supply Chain Improvement Using L-SCOR 1 Introduction 1.1 L-SCOR: An Extension of SCOR for a Self-adaptive Supply Chain 1.2 The Retirement Supply Chain 2 The Retirement Supply Chain Modelling with L-SCOR 3 Implementation and Results Analysis 3.1 Q-Learning as a Reinforcement Learning Algorithm 3.2 State and Action Spaces 3.3 The Reward Strategy 3.4 The Value Function 3.5 The Results 4 Conclusion References Social Media and Information Systems A Hybrid Machine Learning Method for Movies Recommendation 1 Introduction 1.1 Content-Based Filtering Systems (CBF Based Systems) 1.2 Collaborative Filtering-Based Systems (CF-Based Systems) 2 Related Works 2.1 Movie Recommender System Using Single Value Decomposition And K-means Clustering (2020) 2.2 Movie Recommender System Using K-Means Clustering AND K Nearest Neighbor (2019) 2.3 Fully Content-Based Movie Recommender System With Feature Extraction Using Neural Network (2017) 3 Methodology 3.1 Random Forest 3.2 K-Nearest Neighbors 3.3 Singular Value Decomposition (SVD) 3.4 Similarity Computing 3.5 CosSim Similarity 3.6 Mean Absolute Error (MAE) 3.7 Root Mean Square Error (RMSE) 4 Experiments And Results 4.1 Data Set 4.2 Results 5 Comparison 6 Conclusion References On the Sensitivity of LSTMs to Hyperparameters and Word Embeddings in the Context of Sentiment Analysis 1 Introduction 2 Background Information 2.1 Long Short-Term Memory Neural Networks 2.2 Word Embeddings 3 System Design 3.1 The Dataset 3.2 Data Preprocessing and Word Embedding 3.3 The Models Architecture 3.4 Test Scenarios and Performance Metrics 4 Results and Discussion 4.1 The Sensitivity of LSTM and BiLSTM to Batch Size and Epochs 4.2 The Sensitivity of LSTM and BiLSTM to Dropout 4.3 The Sensitivity of LSTM and BiLSTM Word Embeddings Dimension 4.4 Sensitivity of LSTM and BiLSTM to Optimizers 4.5 The Sensitivity of LSTM to Word Embeddings Type 5 Conclusion References Improved Content Based Filtering Using Unsupervised Machine Learning on Movie Recommendation 1 Introduction 2 Outline 3 Related Work 4 Proposed Work 4.1 K-means Clustering Module 4.2 Vector Space Model and User Profile 4.3 Recommendation Module 5 Result and Discussion 5.1 The Dataset 5.2 Implementation 5.3 Evaluation 6 Conclusion and Future Work References The Impact of the k-Nearest Neighbor Parameters in Collaborative Filtering Recommender Systems 1 Introduction 2 Collaborative Filtering Recommender Systems 3 K-Nearest Neighbor Algorithm 4 Experiments 4.1 Evaluation Metrics 4.2 Experimental Evaluation 5 Conclusion References Automatic Detection of Fake News on Twitter by Using a New Feature: User Credibility 1 Introduction 2 Background 2.1 Twitter 2.2 News Features 3 Related Work 4 Our Approach 4.1 New Feature: User Credibility 4.2 Data Preprocessing 4.3 Features Extraction 4.4 Classification Process 4.5 Models 5 Experiments 5.1 Dataset 5.2 Performance Evaluation of Proposed Model 5.3 Data Exploratory 6 Experiments Results 7 Discussion References User-Enriched Embedding for Fake News Detection on Social Media 1 Introduction 2 Introduction 2.1 Fake News Detection on Social Media 2.2 User Representation on Social Media 3 Methodology 3.1 Data Acquisition 4 Implementation and Results 4.1 Exploratory Analysis 4.2 Model Implementation and Results 5 Conclusion References Author Index